Primary Biliary Cirrhosis: estimating resource use and EQ-5D utility from cross-sectional data

by Stephen Rice

Data, glorious data. The precious resource of scientists. Good data are hard to come by, however, and rare diseases naturally suffer from a particular lack of it. Primary Biliary Cholangitis (formerly known as Primary Biliary Cirrhosis (PBC)) is an autoimmune liver disease affecting around 33 to 39 in 100,000 people in the UK and there are significant gaps in knowledge of the disease [1-3]. The UK-PBC study, funded by the MRC, was developed to research the predictive factors of PBC disease, its progression and response to treatment [4 ]. A consortium was established, a network of 150 NHS Trusts, to coordinate the collection of relevant data [5]. Patients were recruited to the UK PBC Cohort from all stages of the disease following diagnosis.

Our part in the project included the estimation of health care costs and patient utility associated with different stages and events in the PBC disease pathway. To this end two postal questionnaires were sent to 2,402 patients recruited to the UK PBC Patient Cohort. One questionnaire focussed on the history of diagnosis and treatment of PBC, its complications and comorbidities. The other focussed on health service resource use, social care resource use, time lost from work and health-related quality of life measured by the EQ-5D-5L . Both questionnaires were returned with sufficient data for an analysis of resource data by 2,236 (93%) respondents, representing roughly 10% of the total UK PBC population. This represents a significant proportion of the UK PBC population.

The cost and utility estimates of PBC obtained from this study are currently the best available for the UK, however there are some limitations to the data. The data collected are assumed to be representative of the entire UK PBC population, but patients at the end of life are likely to be missing from the data set; these patients are likely to incur more costs and have a lower quality of life than the average patient. Other events are also minimally represented in the cohort, such as people waiting for a liver transplant, people who have just had a transplant, and those recently diagnosed. There is, of course, the possibility of recall error as is common in most self-report surveys and resource use questions were limited to the previous 12 months.

Assumptions also had to be made regarding questions that elicited low or inconsistent responses. For example, a multiple-choice question on visits to secondary healthcare workers within the previous 12 months that had mutually exclusive and exhaustive answers requires the respondent to select one of the options, even if there were no visits. If no option was selected, then it was assumed that the respondent had not made any visits to secondary healthcare workers. This is based on the assumption that people are more likely to select one of a series of positive responses than a single negative response. A possible sensitivity analysis is to assume that no response indicates ‘don’t know’ and use some form of imputation. A second example is where the respond selects, ‘no, they haven’t had a drug’, and then proceeds to select a start year and a dose. In this inconsistent case, it was assumed that they did take the drug. It was assumed that a specific positive response would be a more likely indicator of the truth than a negative response.  These assumptions could be investigated by comparing questionnaire responses against officially recorded resource use data, and different question designs could be tested.

Further detail of the study and its results will be available via a full publication soon.

[1] James OF, Bhopal R, Howel D, Gray J, Burt AD, Metcalf JV. Primary biliary cirrhosis once rare, now common in the United Kingdom? Hepatology (Baltimore, Md). 1999 1999/08//;30(2):390-4.

[2] Howel D, Fischbacher CM, Bhopal RS, Gray J, Metcalf JV, James OF. An exploratory population-based case-control study of primary biliary cirrhosis. Hepatology (Baltimore, Md). 2000 2000/05//;31(5):1055-60.

[3] Boonstra K, Beuers U, Ponsioen CY. Epidemiology of primary sclerosing cholangitis and primary biliary cirrhosis: a systematic review. Journal of Hepatology. 2012 2012/05//;56(5):1181-8.

[4] UK-PBC.   [cited 2016; Available from:

[5] Mells GF, Pells G, Newton JL, Bathgate AJ, Burroughs AK, Heneghan MA, et al. Impact of primary biliary cirrhosis on perceived quality of life: the UK-PBC national study. Hepatology (Baltimore, Md). 2013 Jul;58(1):273-83.